Software systems and components may be developed using object technology, and the operation of these systems and components may occur through methods that are performed on and/or by objects. An object's state may be said to include the combination of current attribute values of the object at a particular point in time. The execution of a method may change attribute values of an object, which, in turn, may lead to a new state of the object. Sometimes the current state of the object or computing environment may be an important factor in determining whether a particular action is allowed to be performed or not.
One approach to ensuring that an object performs an action only when allowed by a particular state of the object is programming such requirements into the object itself. Another approach is to rely on the programming of other unrelated objects—that are called by the object to implement all or part of the action—to enforce such requirements.
For example, software that controls an assembly line in a manufacturing plant should be programmed so that a “stop” action should not be performed on the assembly line if the assembly line current is not moving (e.g., as represented by the state of an object representing the assembly line).
Under the first scenario described above, a programmer of the object may directly code this requirement into the object itself, so that when the object receives a “stop” action request, the object checks its own status attributes to make sure that the assembly line is currently moving before allowing the “stop” action to be processed. However, as software projects become larger and more complex, it may become increasingly burdensome for programmers to understand, identify and account for all constraints that are based on the state of an object.
Under the second scenario described above, the programmer of the object may rely on other programming to enforce this requirement. In this example, the assembly line object (which may or may not have its own status attributes regarding the movement of the assembly line) would receive the “stop” active request, and call another unrelated object to implement all or part of the “stop” action. The other object would then check its own status attributes to make sure that the assembly line is currently moving before allowing the “stop” action to be processed, but its determination would be independent of the state of the assembly line object.